ask-chatgptlisted

# Ask ChatGPT ## Overview Ask ChatGPT for one independently useful web result without replacing work Codex or another local owner can complete directly. Codex interprets the user's natural request, fixes the relevant basis or decision, selects the smallest useful theme and ChatGPT capability, supplies boundaries instead of requiring a user-authored prompt, and verifies returned claims or artifacts before downstream use. ## Workflow 1. Read effective guidance and normalize the natural request into the required outcome, subject, known facts, decision or review basis, constraints, exclusions, evidence needs, and stop condition. Ask only when a missing choice would materially change the result. 2. Apply the **Codex-first gate**. If Codex, an existing Skill, or an available host tool can produce the requested result and the user did not request an independent ChatGPT or ChatGPT-web artifact, route there and stop. Use this Skill only when a distinct ChatGPT result is requested or required; never invoke it merely because ChatGPT is available. 3. Freeze one basis type. For Worktree use HEAD plus staged/unstaged patch hashes, in-scope untracked path/content hashes, and recorded exclusions; for immutable review use resolved SHAs; for a decision, research, or creative request record one question, authoritative evidence set, date/version, and exclusions. A decision basis does not require a Git repository. Recheck the selected basis before accepting output. 4. Classify authorization: